Ergodicity, Decisions, and Partial Information
From MaRDI portal
Publication:4568498
Abstract: In the simplest sequential decision problem for an ergodic stochastic process X, at each time n a decision u_n is made as a function of past observations X_0,...,X_{n-1}, and a loss l(u_n,X_n) is incurred. In this setting, it is known that one may choose (under a mild integrability assumption) a decision strategy whose pathwise time-average loss is asymptotically smaller than that of any other strategy. The corresponding problem in the case of partial information proves to be much more delicate, however: if the process X is not observable, but decisions must be based on the observation of a different process Y, the existence of pathwise optimal strategies is not guaranteed. The aim of this paper is to exhibit connections between pathwise optimal strategies and notions from ergodic theory. The sequential decision problem is developed in the general setting of an ergodic dynamical system (Omega,B,P,T) with partial information Ysubseteq B. The existence of pathwise optimal strategies grounded in two basic properties: the conditional ergodic theory of the dynamical system, and the complexity of the loss function. When the loss function is not too complex, a general sufficient condition for the existence of pathwise optimal strategies is that the dynamical system is a conditional K-automorphism relative to the past observations �igvee_n T^n Y. If the conditional ergodicity assumption is strengthened, the complexity assumption can be weakened. Several examples demonstrate the interplay between complexity and ergodicity, which does not arise in the case of full information. Our results also yield a decision-theoretic characterization of weak mixing in ergodic theory, and establish pathwise optimality of ergodic nonlinear filters.
Recommendations
- Games of incomplete information, ergodic theory, and the measurability of equilibria
- On the foundations of decision making under partial information
- Information in Continuous Time Decision Models with Many Agents
- scientific article; zbMATH DE number 3294098
- Maximum probabilities, information, and choice under uncertainty
- scientific article; zbMATH DE number 2143394
- Information-theoretic bounded rationality and -optimality
- Decision making under partial probability information
Cites work
- A complete solution to Blackwell's unique ergodicity problem for hidden Markov chains
- An elementary proof of the strong law of large numbers
- Conditional ergodicity in infinite dimension
- Convergence of empirical processes for interacting particle systems with applications to nonlinear filtering
- scientific article; zbMATH DE number 1713116 (Why is no real title available?)
- scientific article; zbMATH DE number 5871504 (Why is no real title available?)
- scientific article; zbMATH DE number 4153595 (Why is no real title available?)
- scientific article; zbMATH DE number 3745547 (Why is no real title available?)
- scientific article; zbMATH DE number 192938 (Why is no real title available?)
- scientific article; zbMATH DE number 3538599 (Why is no real title available?)
- scientific article; zbMATH DE number 3804533 (Why is no real title available?)
- scientific article; zbMATH DE number 3447235 (Why is no real title available?)
- scientific article; zbMATH DE number 3388896 (Why is no real title available?)
- In general a measure preserving transformation is mixing
- Inference in hidden Markov models.
- Markov Chains and Stochastic Stability
- Mixing Sequences in Hilbert Spaces
- On a class of special flows
- On optimal sequential prediction for general processes
- On the exchange of intersection and supremum of \({\sigma}\)-fields in filtering theory
- Pointwise and {\(L^1\)} mixing relative to a sub-sigma algebra
- Produits tensoriels topologiques et espaces nucléaires
- Some Limit Theorems for Random Functions. I
- The stability of conditional Markov processes and Markov chains in random environments
- The strong law of large numbers for sequential decisions under uncertainty
- The universal Glivenko-Cantelli property
- The Weighted Pointwise Ergodic Theorem and the Individual Ergodic Theorem Along Subsequences
- Uniform Central Limit Theorems
- Uniform time average consistency of Monte Carlo particle filters
- Universal prediction of random binary sequences in a noisy environment
- Weak convergence and empirical processes. With applications to statistics
Cited in
(4)
This page was built for publication: Ergodicity, Decisions, and Partial Information
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4568498)
